No CrossRef data available.
Article contents
EFFICIENCY IN ESTIMATION UNDER MONOTONIC ATTRITION
Published online by Cambridge University Press: 16 September 2024
Abstract
Attrition is monotonic when agents leaving multi-period studies do not return. Under a general missing at random (MAR) assumption, we study efficiency in estimation of parameters defined by moment restrictions on the distributions of the counterfactuals that were unrealized due to monotonic attrition. We discuss novel issues related to overidentification, usability of sample units, and the information content of various MAR assumptions for estimation of such parameters. We propose a standard doubly robust estimator for these parameters by equating to zero the sample analog of their respective efficient influence functions. Our proposed estimator performs well and vastly outperforms other estimators in our simulation experiment and empirical illustration.
- Type
- ARTICLES
- Information
- Copyright
- © The Author(s), 2024. Published by Cambridge University Press
Footnotes
We are very grateful to the Editor (Peter C. B. Phillips), the Co-Editor (Patrik Guggenberger), two anonymous referees, and Whitney Newey for their help in improving the paper. We also thank Francesco Amodio, Marine Carrasco, Daniel Farewell, Bryan Graham, Fabian Lange, Steven Lehrer, Thierry Magnac, Erica Moodie, Chris Muris, Tom Parker, Geert Ridder, Youngki Shin, and various conference and seminar participants for helpful comments. Earlier versions of the paper were circulated under different names; e.g., “A note on efficiency in estimation with monotonically missing at random data.” The views presented in this work do not reflect those of Analysis Group. Analysis Group provided no financial support for this work.